Causal Inference on Neuroimaging Data with Mendelian Randomisation

Abstract

While population-scale neuroimaging studies offer the promise of discovery and characterisation of subtle risk factors, massive sample sizes increase the power for both meaningful associations and those attributable to confounds. This motivates the need for causal modelling of observational data that goes beyond statements of association and towards deeper understanding of complex relationships between individual traits and phenotypes, clinical biomarkers, genetic variation, and brain-related measures of health. Mendelian randomisation (MR) presents a way to obtain causal inference on the basis of genetic data and explicit assumptions about the relationship between genetic variables, exposure and outcome. In this work, we provide an introduction to and overview of causal inference methods based on Mendelian randomisation, with examples involving imaging-derived phenotypes from UK Biobank to make these methods accessible to neuroimaging researchers. We motivate the use of MR techniques, lay out the underlying assumptions, introduce common MR methods and focus on several scenarios in which modelling assumptions are potentially violated, resulting in biased effect estimates. Importantly, we give a detailed account of necessary steps to increase the reliability of MR results with rigorous sensitivity analyses.

Genetic architecture of brain age and its casual relations with brain and mental disorders

“…It is also worth noting that while the sample sizes for the disorders are large, the one underlying our BAG GWAS is relatively small. Thus, our null ndings in the direction from BAG to disorders may be due to too weak instruments 70 .…”

Genetic architecture of brain age and its casual relations with brain and mental disorders

“…It is also worth noting that while the sample sizes for the disorders are large, the one underlying our BAG GWAS is relatively small. Thus, our null findings in the direction from BAG to disorders may be due to too weak instruments 70 .…”

Statistical learning methods for neuroimaging data analysis with applications

“…Causality research has received a lot of attention in neuroscience research (Friston, 2009;Ramsey et al, 2010;Lindquist, 2012;Yu et al, 2022;Sobel and Lindquist, 2020;Yu et al, 2022;Taschler et al, 2022;Knutson et al, 2020;Le and Stein, 2019;Zhao and Castellanos, 2016;. Some important scientific questions in neuroscience include how experi- Causality research can be roughly divided into causal discovery for determining causal relationships among a set of variables and causal inference for estimating causal effects deriving from a change of a certain variable over an outcome of interest in a large system (Imbens and Rubin, 2015;Pearl, 2009;Greenland et al, 1999;Upadhyaya et al, 2021;Imbens, 2020).…”